Pharmacokinetic models for the saturable distribution of paclitaxel

Paclitaxel pharmacokinetics are nonlinear with saturable metabolism and saturable distribution to the tissues. The saturable distribution has in previous pharmacokinetic modeling been described as a saturable transport process, whereas the present study was undertaken to investigate alternative explanations. Using a sparse sampling scheme (on average 3.3 samples per profile), 101 plasma concentration-time profiles in 22 female patients with metastatic cancer of the breast or ovary were monitored. It was found that the observed data could be equally well described by saturable tissue binding as well as by capacity-limited tissue transport. The data were better described by a model where equilibrium was achieved with drug in the central rather than in the peripheral compartment. Models where the binding was assumed to be an instantaneous or a noninstantaneous process were tried, but the data did not allow resolution between these two possibilities. The value at which the saturable transport was half-maximal was 0.55 microM. The K(d) values of the binding models were 0.06 to 0.12 microM. These are close to the values reported as a threshold for drug toxicity of paclitaxel, suggesting a possible connection between the binding sites involved in the pharmacokinetics and the mechanism responsible for the toxicity. For all models, a saturable elimination of paclitaxel was included using the Michaelis-Menten model. K(m) for the elimination ranged in the different models from 2.5 to 5.6 microM.